Finally... Uniform Initialization FTW!!!!!!!!!!! Can't wait for Initializer lists to make it in at long last!!! Also STL if you want a good editor suggestion, my favourite & recommendation = EmEditor http://www.emeditor.com/ , it uses boost shows it proudly in "About"
It's a 64-Bit editor which has multiple very sweet properties, Visual Studio Project Plugin and tons of more plugins, snippets for countless languages, auto-completion for... everything even based on previous files used, uses ZERO (0) CPU cycles when not in use AND it can handle many-millions of lines of text without slowing down even doing vertical editing on line *million .
I think they've got a slightly wrong/too damn conservative mindset here. Sure jitting methods when their not needed may be a "waste" in terms of I want to flash deep fry this entire buffalo in 10 seconds flat. But I believe Jitting in chunks when they seem probable to be used is a far better way to go and cache the jitted methods for use later on, ambient CPU and RAM size is still growing...
A good analogy to this is the CPU's themselves, not like the CPU grabs a single friggin bit of data out of memory, the CPU's pull the data as cache-lines and chunks at a time for faster speeds even tho the data in the memory isn't all being used. Of course it causes things like false sharing and memory location contention when it's not done right but still, hardware went to grabbing more data than is needed in one go, I think that Vance shouldn't be so picky on being 100% efficient as if it were kernel space or if this code was used for lunar launches & space exploration. If he wants, why don't they just put in heuristics or hints to the CPU's so that CPU's can also be 98% efficient using/hinting to the branch prediction engines built into current CPU's and do their tracing to help it along?
@Charles:You did a fine job as always Charles!! don't sweat it... it just wasn't what I was expecting Somehow things went in unusual directions on their own even with your moderation and productive input throughout.
I'm not going to bother giving a long comment as any dev worth their salt would have already been disappointed... But I will say that when I saw these two together I got giddy and was expecting nothing less than a masterpiece of conversation. Somehow it turned into a church-vs-church fiasco. Wish they would've gotten deeper than tree-shaking and actually spent time on matters which actually defined their design decisions and helped one another understand each others essential conceptual logic for why their solutions were chosen and different by design.
If you want to see the slides until Charles works something out (as I honestly don't care to see Erik and is far more interested in what he's saying + the presentation) open one of the video file links in VLC and use "Effects and Filters" on the video to crank up the contrast and lower the brightness to see the slides quite clearly
Thank you for another Going Native episode, along with STL's C++ series' are now the BEST and most enjoyable videos on C9 to me since Going Deep isnt exactly interesting lately focusing only on Async/Rx with a lack of Erik/Jim Radigan.
@Hakime:LLVM/Clang uses a runtime ... http://compiler-rt.llvm.org/ learn and weep ALL native program code other than OS Kernel code specifically developed to be used inside the kernel uses a "runtime", and even then, since I don't have access to Windows kernel code it probably also creates it's own "runtime" at its very core, but it's all self-contained and doesn't have any external dependencies to classify the kernel's core as a "runtime".
DevDiv: The closer you stick to implementing C++11 to it's official specification, the better you will be. And PLEASE focus on the CORE LANGUAGE before you spiff it up with the flashy libraries you mentioned. I'll love seeing it complete to the best of Microsoft's ability and finally once again be the best compiler available from day's long past.
And I'm going to say... I actually dislike the way AMP was implemented... It's results are impressive but the way you focus on array_view's just feels... bad... and yes OpenCL's Warp focused structure feels bad too
I love the Windows 8 UI but the most stupid decision MS has made regarding UI was how buried/unusual/different shutting down the PC is, by using the "Setting" charm, To me it's the only negative point about Metro, if they remedy that by BY DEFAULT put the shutdown option in a more accessable place they would have done Metro absolutely beautifully.
BTW C'mon Charles lets get some WinRT architectural video's/Win8 videos out!.
This guy was amazing in the previous video and I'll be downloading this now to watch asap!
I wonder what Jim(or anyone who's interested here) would think of my parallel execution engine/vm which schedules and tracks data dependance in a sort of VM environment with a 1 extra layer of memory location indirection instead of making the compiler know all data & locations aware at compilation time?
It is a little vague and it does it's break up horizontally accross cores instead of vertically to fill the vectors but I did try to explain for a broader audience but I'd love feedback on if my architecture & design seems possible! If it's just wrong or if any of you long time researchers know if it can't work as I tried to explain please let me know .